Computation of Optimal Auctions: An efficient implementation of computing revenue-optimal auctions via reduced forms

Abstract

While little is known about revenue-optimal auctions in a multi-bidder, multi-item setting, computational approaches are able to provide some insights into their structures. Previous work by Cai, Daskalakis, and Weinberg presents an efficient way of calculating such auctions by solving a linear program with a separation oracle, using reduced forms to make the problem more tractable. In this thesis, we present RoaSolver, a software program that implements these ideas.